Novel insights into kidney disease: the scRNA-seq and spatial transcriptomics approaches: a literature review

Abstract

Over the past decade, single-cell RNA sequencing (scRNA-seq) and spatial transcriptomics (ST) have revolutionized biomedical research, particularly in understanding cellular heterogeneity in kidney diseases. This review summarizes the application and development of scRNA-seq combined with ST in the context of kidney disease. By dissecting cellular heterogeneity at an unprecedented resolution, these advanced techniques have identified novel cell subpopulations and their dynamic interactions within the renal microenvironment. The integration of scRNA-seq with ST has been instrumental in elucidating the cellular and molecular mechanisms underlying kidney development, homeostasis, and disease progression. This approach has not only identified key cellular players in renal pathophysiology but also revealed the spatial organization of cells within the kidney, which is crucial for understanding their functional specialization. This paper highlights the transformative impact of these techniques on renal research that have paved the way for targeted therapeutic interventions and personalized medicine in the management of kidney disease.

Introduction

Kidney diseases represent a significant global health challenge due to their complex etiology which includes genetic, environmental, and lifestyle factors [1]. Surveys have indicated a rising prevalence of kidney disease that often occurs at increasingly younger populations [2, 3]. Acute kidney injury (AKI) is become a common complication among hospitalized patients, particularly those who are critically ill, and is associated with both short- and long-term mortality [4]. Chronic kidney disease (CKD) affects 8–16% of the global population, with a prevalence of 8.2% in China, highlighting the importance of early detection, diagnosis, and intervention to mitigate its effects [5, 6].

Over the past decade research on kidney disease has seen substantial progress driven by single-cell RNA sequencing (scRNA-seq) technology and spatial transcriptomics (ST) [7,8,9]. The development of these technologies has advanced biomedical research significantly, particularly in elucidating cellular heterogeneity in kidney disease. In contrast, the emergence of scRNA-seq technology has allowed scientists to measure gene expression at the single-cell level, thereby providing insights into previously unknown subpopulations of cells in kidney disease and their roles in disease progression. ST technology has further expanded our horizons by combining histopathologic and molecular analyses to provide detailed information on cell location and gene expression. This technology not only identifies cell types but also reveals cell-cell interactions and their functions within the tissue microenvironment. For instance, ST techniques have enabled researchers to observe at the tissue level how cells respond to acute injury, participate in kidney development and disease processes, and also their spatial distribution and interactions in pathological states [10,11,12]. These technological advances offer new perspectives for the diagnosis, treatment, and prevention of kidney diseases. They have helped identify potential new therapeutic targets and may pave the way for personalized medicine. The aim of this paper was to review the current state of scRNA-seq and ST in the context of kidney disease research. We will examine how these techniques have transformed our understanding of renal cellular heterogeneity, the molecular mechanisms underlying kidney function and disease, and the potential for these insights to inform future therapeutic strategies.

scRNA-seq technology

scRNA-seq is a technology that enables the analysis of the transcriptome of individual cells, revealing the heterogeneity between cells and the gene expression patterns of specific cell types. scRNA-seq can display the gene expression profiles of each cell, aiding in the identification of new cell subtypes, cell states, and interactions between cells. In kidney disease research, scRNA-seq is widely used to reveal the trajectories of cellular changes, characteristics of cell differentiation, and therapeutic targets during the occurrence and development of diseases [13].

Over the past decade, the development of new cell isolation and sequencing techniques has enabled to conduct whole genome RNA analysis of individual cells using scRNA-seq techniques [14]. Compared to conventional research methodologies which focus primarily on evaluating mean expression levels across a substantial quantity of cells or tissues, scRNA-seq offers a novel approach that identifies rare cells, subtypes of cells, disease-specific cell types as well as analyze cell-to-cell interactions through ligand-receptor interactions [15,16,17,18] (Fig. 1). During the scRNA-seq process, tissue is first lysed into a suspension of dispersed viable cells, with single cells then isolated for the RNA sequencing procedure. In addition, the development trajectory of different types or subtypes of cells can be tracked and analyzed by computational analysis, such as pseudotime analysis [19] (Fig. 1). The continuous updating and development of single-cell sequencing technology have made it an increasingly important tool in analyzing cell heterogeneity, revealing the relationship between cell populations in the microenvironment, and tracking the occurrence and development of diseases [20,21,22].

Fig. 1

Kidney tissue is dissociated into a single cell suspension and scRNA-seq is then performed. ScRNA-seq allows the study of rare cell types, cell state and subtype heterogeneity, disease-specific cell types, and cell-to-cell interactions via ligand-receptor analysis. Computational analyses such as pseudo-time diffusion mapping analyze the similarity and diversity of cells, consent to trace differentiation processes, clonal evolution, and cell state transitions between different cell types

To initiate the renal scRNA-seq assay, the first steps involve proper lysis of tissue and the preparation of a monodispersed viable cell suspension. This can be achieved through a variety of methods, including fluorescence-activated cell sorting (FACS), separation by magnetic beads of specific antibodies, the chip-based low-pressure microfluidic technique, microdroplet platforms, and techniques of automated micromanipulation or laser capture microdissection [23] (Fig. 2). Flow cytometry is one of the most widely used methods for simultaneously evaluating cell size, viability, and aggregation, as well as identifying specific cell subsets. Microfluidic technology using a high throughput transient droplet microfluidic technique is used widely due to its high capture efficiency and a low cost [24]. By separating cells into individual water-in-oil droplets, microfluidic technology uses a continuous flow of oil to bind single cells with magnetic beads and primers containing unique bar codes [25]. After isolation of single cells, the scRNA-seq library is generated following cell lysis, followed by reverse transcription into complementary cDNA, amplification by polymerase chain reaction (PCR), or in vitro transcription and in-depth sequencing. Introduction of unique molecular identifiers (i.e., UMIs or bar codes) before cDNA amplification improves data accuracy by eliminating amplification bias generated by the polymerase chain reaction. To date, a variety of single cell sequencing techniques have been developed that can be grouped roughly into two types, full-length type and tag-based type, mainly represented by Smart-seq2 and Drop-seq. Smart-Seq2 is a method for reverse amplification of a full-length transcript of a single cell using templates, that was previously limited in application due to its high cost, but is now used widely in a variety of scRNA-seq after modifications [26,27,28,29]. In terms of performance, Smart-seq2 has higher sensitivity, while Drop-seq reduces the cost. In different droplet-based microfluidic techniques, 10XChromium is known for its higher sensitivity and lower error rates [14]. The combination of two or more scRNA-seq techniques may increase the possibility of capturing rare cell types and transcripts with a low abundance. Therefore, selection of the scRNA-seq platform should be guided by the specific objectives and aims of the study, with the appropriate combination of techniques chosen according to the study design and the desired endpoint.

Fig. 2

Main steps in the scRNA-seq workflow. First, the tissue of interest is dissociated to make a single-cell suspension. Single cells are then harvested for scRNA- seq analysis. Magnetic activated cell sorting relies on the immunoreactivity of cell specific antigens with magnetic beads. A fluorescence activated cell sorting platform then selects individual cells with heterogeneous tissue by detecting fluorescent labelled signals. The cells can be isolated using a variety of parameters. Smart-seq2 and CEL-Seq2 are performed in 96 or 384-well plates after sorting, while droplet systems (e.g. 10X Chromium and Drop-Seq) couple cells with barcoded beads containing a unique molecule identifier (UMI) and primers that form water in-oil droplets via a continuous oil flow. Reverse transcription and cDNA amplification are performed by PCR in Smart-Seq and 10X Chromium and by in vitro transcription (IVT) in CEL-Seq2

Application of scRNA-seq technology in renal research

Identification of cell types and subpopulations in renal tissue

Single-cell sequencing has been used to investigate gene expression patterns and the functions of different types of cells in the kidney. This technology enables researchers to identify and classify various cell subtypes in the kidney, such as tubular cells and mesangial cells, and also investigate their roles and interactions in kidney diseases (Fig. 1). For example, a 2021 study identified heterogeneity in mesangial cells in mouse glomeruli, including four cellular subpopulations of Pdgfr + mesangial cells [30]. Furthermore, Liu et al. [31] used single-cell sequencing to identify five distinct subpopulations of glomerular epithelial cells (PECs) labeled as PEC-A1, PEC-A2, PEC-A3, PEC-A4, and PEC-B, which included the presence of podocyte progenitor cells in PEC-A1 and PEC-A2 and tubular progenitor cells in PEC-A4. Importantly, dynamic signalling network analysis highlighted the pivotal role of PEC-A4 activation and PEC-A3 proliferation in crescent formation. Insights gained from these scRNA-seq-based analyses have contributed to our understanding of the pathology of crescentic glomerulonephritis and suggested potential therapeutic strategies [31]. With single-cell sequencing, researchers are able to obtain the gene expression profiles of individual cells in the glomerulus, thereby revealing the functions and regulatory mechanisms of these different cell types in physiological and pathological states.

Similarly, single-cell studies have demonstrated various subpopulations of cells within the proximal tubule, both in healthy and damaged proximal tubules, suggesting significant heterogeneity [32,33,34]. Furthermore, recent studies have characterized two distinct cellular subpopulations of thick ascending limb in the human kidney using single-nucleus assay for transposase-accessible chromatin (snATAC) and RNA (snRNA) sequencing, that suggested the presence of plasticity among principal cells (PCs) and intercalary cells (ICs) in the collecting ducts [35]. Single-nucleus RNA sequencing (snRNA-seq) is a sequencing technology that extracts RNA from individual cell nuclei, particularly suitable for tissue samples where intact cells are difficult to obtain. This technique excels in revealing gene expression information within the cell nucleus, thus providing a vital tool for studying transcriptional regulatory mechanisms inside the nucleus. Compared to traditional scRNA-seq, snRNA-seq demonstrates greater flexibility and effectiveness in handling frozen samples and tissues that are hard to dissociate [36]. Single-cell transcriptome analysis of healthy human kidneys also identified sex differences in gene expression in proximal tubular cells, with females showing increased expression of antioxidant metallothionein genes and males having increased expression of aerobic metabolism-related genes. Metabolic functions in proximal renal tubular cells also vary between the sexes, with male cells displaying higher levels of oxidative phosphorylation and energy metabolite precursors [37]. In addition, the study identified kidney-specific lymphocyte subpopulations with unique transcriptional profiles and revealed significant heterogeneity in medullary-derived cells that indicated the predominant cellular population in the healthy kidney as MRC1 + LYVE1 + FOLR2 + C1QC + cells [37]. Furthermore, scRNAseq data indicated that there were four major subpopulations of monocytes and two major subpopulations of kidney-resident macrophages (KRMs) in healthy mice [38]. These findings collectively provide important insights for understanding renal development, disease occurrence and progression, and highlight the potential of single-cell sequencing technology as a powerful tool for identifying rare cells and guiding the identification of new therapeutic targets for kidney diseases. As a consequence, the continuous development of single-cell sequencing continues to provide new insights into physiological and pathological mechanisms in the kidney.

Identification of interim cell types or potential progenitor cells in renal development and regeneration

Single-cell sequencing technologies are also widely used to study cellular dynamics during kidney development and regeneration. By analyzing kidney tissue during human embryonic development, researchers have identified several key cell subpopulations, including renal precursor cells and mesenchymal stem cells (Fig. 3A). These cell subpopulations play important roles in the formation of renal units and maintain normal kidney development through dynamic homeostasis [37, 38]. In addition, studies have revealed the potential roles of these cell subpopulations in congenital kidney disease, providing new perspectives for understanding the pathogenesis of related diseases [39].

Fig. 3

The focus of spatial transcriptomic research. (A) “Kidney development” refers to the study of how the spatial transcriptome changes during key stages in kidney development; (B) “Kidney homeostasis” refers to elucidating spatial division of discrete cellular subtypes in healthy renal tissue at a singular time point; (C) “Kidney injury microenvironment” refers to elucidating the spatial transcriptome in injured tissue niches in relation to their proximity to relevant biological traits; (D) Integration of single-cell and ST data

scRNA-seq can be used to investigate rare cell types (Fig. 1). For example, using the data of single-cell sequencing analysis, Nie et al. [39] identified seven cell populations in healthy human urine and suggested that SOX9 + cells might be potential progenitor cells. Muto et al. [35] also used single-cell transcriptome and chromatin accessibility data to identify a new cell subtype in proximal tubules named PT_VCAM1. Further study revealed that CD24 and CD133 were expressed in VCAM1 + cells, suggesting that VCAM1 + proximal tubule cells had progenitor cell-like characteristics. In the ischemia-reperfusion injury (IRI) mouse model, the proportion of PT_VCAM1 + cells was calculated after deconvolution of RNAseq, with the results showing that the number of cells had increased significantly 24 h after IRI that lasted for at least 7 days. These findings indicate that PT_VCAM1 + cells may represent a type of regeneration-associated cell in the proximal tubules. Hilliard et al. [40] also used single-cell transcriptomics to determine the mechanism by which renal progenitor cells stopped generating new filtration units when mouse embryonic kidneys differentiated into mature kidneys. This research integrated open chromatin domains, represented promoters and enhancers, with gene expression in the same single cells, and then characterized the composition and DNA of regulatory gene functions in these cells. The results showed that multi-genomics identified the sequence characteristics of the “precursor” bHLH/Fox gene in renal progenitor cells and expression of Foxp1 which played a key role in kidney formation by increasing podocyte chromatin. In this research, specific sequence motifs within the bHLH/Fox gene family were identified, which are recognized as “pioneer” genes in nephron progenitor cells. This study particularly highlights the important role of Foxp1 in modulating podocyte development. It is noteworthy that genes implicated in renal cell proliferation are transcriptionally inactivated postnatally, yet they retain the capacity for activity within the genomic landscape. A recent single-cell analysis revealed a heterogeneous activation pattern in regenerated hepatocytes, with pro-regeneration genes remaining in the active chromatin state, but marked as DNA methylome and tri-methylation at lysine 27 of histone H3 (H3K27me3) and silenced in the quiescent liver. During regeneration, the H3K27me3 promoter was depleted, which facilitated the activation and dynamic expression of the pro-regeneration gene [41]. Whether or not cell regeneration after kidney injury depends on the regulation of H3K27me3 remains to be further clarified. Regeneration is the key to repair after kidney injury and how to identify progenitor cells with renal regeneration ability is essential. Taken together, the results of these single-cell sequencing studies provide insights into the transformation of cell types and regeneration-related gene expression in the injured region. Single-cell sequencing can also aid in the identification of potential regeneration-related cell subpopulations and signaling pathways during the recovery period from kidney injury. The SOX9 + progenitor cells isolated from the urine of patients with chronic kidney disease were cultured and transplanted into the mice with injured kidneys, that demonstrated human renal progenitor cells could bind with and repair damaged kidney tissue of the mice. This finding suggests the potential for using autologous cell transplantation to treat kidney disease in the future [39]. As shown in Fig. 3B, sc-RNA seq techniques can be used to identify different cell types in developing kidneys, thereby revealing specific cell types and transcriptomic changes during kidney development. This technology also helps researchers identify potential stem cells, precursor cells, and regeneration-associated cell subpopulations. Valuable information on kidney development and disease pathogenesis can be gained by accurately characterizing conserved and mutated features at key stages of human kidney development and injury, as well as identifying the drivers of regeneration in disease states. By using a full-length sc-RNA seq technique, Wineberg [42] studied alternative splicing in mouse embryonic kidneys and comprehensively elucidated a key conversion process in kidney development that involved splice isomer conversion in the conversion between mesenchymal and epithelial cells. That study also identified Rbfox1/2 and Esrp1/2 as the splicing regulators of the hepatocyte growth factor receptor (MET) in renal development, a finding which improved the understanding of the molecular mechanisms that drive renal development. This knowledge will also help to further investigate repair mechanisms and identify potential cell types involved in kidney regeneration after injury.

Kidney and renal disease models from a single-cell perspective

The renal microenvironment, including renal intrinsic and extrinsic cells (mainly immune cells), plays a significant role in the pathogenesis of all acute and chronic kidney diseases [43]. In response to chronic kidney injury, immune cells secrete pro-inflammatory cytokines and chemokines as well as activating quiescent fibroblasts to acquire myofibroblast phenotypes which produce fibrous matrix proteins, destruct tissue structure, and contribute to fibrosis. This renal immune disorder is common in different types of CKD and triggers a series of consequences such as cell stress and death, infiltration of inflammatory cells, apoptosis, myofibroblast activation, proliferation of kidney intrinsic cells, tubular atrophy, microvascular rarefaction, and renal fibrosis [44] (Fig. 3C).

Identification of renal macrophage subtype in acute kidney injury and chronic kidney disease

The highly complex function of the kidneys maintaining the homeostasis of the body involves continuous production of urine to excrete metabolic wastes and excess water, regulate electrolytes, acid-base balance, and blood pressure, as well as secreting hormones such as erythropoietin and active vitamin D. The maintenance of normal renal function requires the interaction between many different types of cells with a high degree of specificity [45,46,47]. In-vitro and in-vivo studies have shown that macrophages actively participate in the early injury and repair process of AKI inflammation and therefore have become a potential therapeutic target [48,49,50]. Research by Conway et al. [18] created transcriptional expression profiles of immune cells in the kidney injury-repair process using single-cell sequencing technology in the unilateral ureteral obstruction (UUO) and reverse UUO (R-UUO) mouse models. The study demonstrated that in the early stage of urinary tract obstruction injury, Ly6c2 + and Arg1 + mononuclear cells showed an aggregation trend in the kidney and expressed genes related to hypoxia, inflammation, and fibrosis. After removal of the urinary tract obstruction, a new macrophage subpopulation, MMP12 + was identified, which had the effect of reversing fibrosis, indicating a potential new therapeutic target for CKD. Results from the IRI-AKI mouse model showed that the majority of macrophages in the kidney at the early stage injury were derived from blood mononuclear cells, and that they could be divided mainly into two subsets, each with distinct functions. In addition, small molecule inhibitors targeting the S100a9hiLy6chi signaling pathway were found to play a protective role by reducing the inflammatory response in the bilateral and unilateral IRI models, suggesting that S100a9 could be a potential drug target for the treatment of AKI [51]. Another research team investigated the functional and positional heterogeneity of KRMs in bilateral ischemic kidney injuries using sc-RNA seq and identified seven different KRM subsets, each with unique transcriptomic characteristics [51]. Recent studies have also reported that immune cells undergo a second activation after U-IRI, resulting in aggravated renal tubular atrophy induced by inflammation [52]. Immune cell elimination significantly ameliorates AKI-induced renal tubular atrophy and inhibits the conversion of AKI to CKD. The changes in the localization and distribution of KRMs confirm the long-standing hypothesis of local immune imbalance after AKI and explain the possible causes of the increased risk of CKD after AKI [53].

Increasing evidence demonstrates that renal macrophages play a significant role in the development of diabetic kidney disease (DKD). In DKD lesions, macrophages are the main inflammatory cells, which can be divided into various subtypes [54]. Fu et al. [55] performed single-cell sequencing on early DKD mice and showed that the proportion of macrophages exceeded the sum of other immune cells. Further immunofluorescence staining of human kidney biopsies revealed a marked increase in the number of MRC1 + and TREM2 + infiltrating macrophages in patients with DKD, which was consistent with the results in a DKD mouse model. This finding suggested that macrophages are involved in the regulation of renal inflammation at the early stage of DKD. In the early stage of DKD, expression of IFNhi macrophage (Mac), Mrc1hi Mac, and Trem2hi Mac were shown to be increased significantly. Recent studies have shown that the P2Y12 receptor is highly expressed in the macrophages of both CKD patients and UUO mouse models, and promotes the transformation of macrophages into myofibroblasts (macrophage-myofibrolast transition, MMT) and the development of renal fibrosis through TGF-β1/Smad3. The application of P2Y12 inhibitors blocks MMT and effectively inhibits renal fibrosis, providing a new approach and theoretical basis for the treatment of CKD renal fibrosis [56]. Wu et al. [57] performed sc-RNA seq on healthy and transplanted kidney biopsies and investigated the role of donor macrophages in kidney transplant rejection. The study showed that the recipient macrophages exhibited inflammatory activation, while the donor macrophages exhibited antigen presentation and complement signal transduction. The leukocyte donor/receptor ratio was also observed to vary according to the state of macrophage rejection, with donor-derived macrophages having different transcriptional profiling compared to that of recipient macrophages. Collectively, these findings indicated that immune cell chimerism participates in rejection after renal transplantation [47].

In summary, these results indicate that macrophages play an important role in the progression of AKI and CKD. Kidney-resident macrophages and mononuclear phagocytes derived from the circulatory system can influence the fate of damaged kidneys. Current research is investigating the methods of targeting macrophage subsets to obtain therapeutic effects, although more research data are needed to fully understand the roles of various new macrophage subtypes in different etiologies of acute and chronic renal diseases.

Revealing multiple cell-cell interactions between renal intrinsic cells and immune cells

IgA nephropathy is characterized by the deposition of IgA in the glomerular mesangial region. Renal biopsy pathology shows that macrophages and T cells are the main types of infiltrating cells in the IgAN mesenchyme. Single-cell sequencing has demonstrated that in-situ macrophages in the IgAN kidney exhibited abnormalities in biological processes, such as the NOTCH signaling pathway, glycolysis, and fatty acid metabolism. Moreover, the high expression of JCHAIN in mesangial cells suggests that JCHAIN may be a significant factor why the IgA antibody immune complex tended to deposit in the mesangial region [58]. The application of single-cell sequencing technology has helped to gain a deeper understanding of the onset and progression of IgA nephropathy by resolving the functional abnormalities of different cell populations and discovering new signaling pathways. Single-cell sequencing analysis also helped to identify abnormal cell types and new signaling pathways, which in turn guided the development of more precise therapeutic strategies. In addition, single-cell transcription differential analysis showed that the differentially expressed genes of IgAN mesangial cells were enriched mainly in items related to cell adhesion and extracellular matrix. Further cellular communication analysis showed that the mutual communication between mesangial cells and endothelial cells is particularly strong, indicating that the close interaction between these cells may contribute to the pathological changes in the IgAN mesangial region [59]. A single-cell study of molecular changes in renal endothelial cells in ddY IgAN mice focused on the characteristics of immune cell recruitment and infiltration induced by endothelial cells and cell-to-cell interaction in the early stage of IgA nephropathy. This study demonstrated that cross-talk between endothelial cells and immune cells, and the possible existence of cross-talk between endothelial cells, mesangial cells, podocytes, and other intrinsic cells of the glomerulus were jointly involved in the development of early IgAN glomerular injury [60] (Fig. 3D).

Revealing the origin cells and interaction network in renal fibrosis

The cellular origin of renal fibrosis has been extensively investigated in recent years, with scRNA-seq technologies providing critical insights into the heterogeneity and dynamics of fibrogenic cell populations [61]. Accumulating evidence supports a predominant role of resident mesenchymal cells, particularly PDGFR-β-positive pericytes and fibroblast subpopulations, as the primary source of scar-forming myofibroblasts in chronic kidney disease (CKD). These cells undergo progressive differentiation into ECM-secreting myofibroblasts under sustained activation of profibrotic pathways such as TGF-β, Wnt, and PDGF-α, as demonstrated by Kuppe et al. [62] through pseudotemporal trajectory analysis and multicolor in situ validation. Notably, scRNA-seq data further highlight the pivotal contribution of tubulointerstitial crosstalk, wherein injured renal tubular epithelial cells exhibit aberrant activation of Notch, TGF-β, and Wnt signaling, driving a pro-fibrotic microenvironment. This phenomenon aligns with the identification of VCAM1 + tubular cells in both acute and chronic kidney injury models, which secrete profibrotic mediators to amplify mesenchymal cell activation [63].

While circulating monocytes exhibit minimal direct ECM production [64], emerging studies on transplant-associated renal fibrosis reveal their indirect role in modulating fibrogenesis. Transcriptomic profiling of renal allografts demonstrates that immune-parenchymal interactions (e.g., endothelial-mesangial and epithelial-immune crosstalk) activate profibrotic pathways, including the Notch3-IL-6-integrin axis [65]. Notch3 signaling, in particular, exerts dual effects: it promotes ECM deposition via NF-κB-mediated upregulation of Sonic Hedgehog (Shh) and Tenascin-C, while concurrently enhancing immune cell recruitment. These findings position Notch3 as a promising therapeutic target for CKD [66]. Additionally, scRNA-seq has uncovered functional heterogeneity within glomerular mesangial cells, where PDGFRβ + SCARA5 + subpopulations contribute to fibrosis through paracrine signaling, corroborating earlier hypotheses regarding mesangial cell involvement [64].

A key unresolved issue lies in context-dependent variability of cellular contributions across disease models. Resident mesenchymal cells dominate in CKD, whereas acute injury or transplant-related fibrosis involves dynamic interplay between tubular epithelial dedifferentiation, immune cell infiltration, and mesenchymal activation. Recent stratification of renal allografts into low- and high-ECM states via transcriptomic analysis has advanced our understanding of fibrotic heterogeneity, yet the absence of longitudinal tracking limits mechanistic clarity regarding temporal transitions in cell fate [65]. Future research must integrate spatiotemporal single-cell profiling, organoid-based modeling, and clinical biomarker discovery to delineate stage-specific roles of distinct cell lineages.

Single-cell sequencing technology in the treatment of kidney disease

Single-cell sequencing technology has the potential to enhance our understanding of disease mechanisms and facilitate the development of personalized therapy for kidney disease. Wang et al. [67] used single-cell sequencing to investigate kidney cells of DKD mice treated with SGLT2i, ARB, or a combination of both, with the aim of determining the specific mechanisms underlying the renal benefits of these drugs. Their findings indicated that ARBs primarily exerted their renoprotective effects by mitigating inflammation and fibrosis, while SGLT2i primarily affected mitochondrial function to achieve renoprotective effects. Another study [68] used sc-RNA seq to investigate the response of a mouse model of DKD to five different treatment regimens. The study mapped the single-cell transcriptomes of nearly one million cells, revealing heterogeneity among different renal cell types and variability in their response to the DKD environment and different treatment regimens. In addition, the study highlighted the potential benefits of combination drug therapy over single drug therapy. In conclusion, single-cell sequencing technology offers valuable insights into the cellular heterogeneity of the renal transcriptional response to DKD and standard therapeutic regimens, offering new perspectives on the pathogenesis and treatment of DKD.

Wang et al. [67]. conducted a study using scRNA-seq to investigate the diversity of transcriptomes in renal tubular epithelial cells (TECs) and immune cells. The main focus of the study was to explore the potential mechanisms of umbilical cord MSCs in treating AKI caused by ischaemia-reperfusion injury. The findings revealed that miR-26a-5p present in MSC-derived exosomes played a role in reducing the proportion of profibrotic TECs and the infiltration of immune inflammatory cells by targeting and inhibiting the expression of Zeb2. As a consequence, this intervention contributed to decreased progression of renal fibrosis. In summary, the findings of this study provide a theoretical basis for the clinical application of MSC therapy in the treatment of AKI.

ST

ST is an emerging technology that integrates traditional histology with high-throughput RNA sequencing to explore spatial gene expression patterns in cells or tissues [69, 70]. The core advantage of ST is its ability to provide spatial distribution information of gene expression within cells and tissues, which is critical for understanding cellular phenotypes and states [71]. Unlike scRNA-seq, which involves the dissociation of cells into a suspension and consequently results in the loss of spatial information, ST preserves this spatial context by analyzing intact tissues [72]. This preservation of spatial information is essential for studying the relative proximity of cells and their interactions with surrounding non-cellular structures [73].

Spatial transcriptomics technologies have seen remarkable advancements, leading to significant breakthroughs in molecular biology. Initially, in situ hybridization techniques like SISH (Single In Situ Hybridization) provided spatial information on gene expression but were limited in resolution, restricting insights at the cellular level [74]. The development of multiplexed techniques such as MERFISH (Multiplexed Error-Robust Fluorescence In Situ Hybridization) and seqFISH (Sequential Fluorescence In Situ Hybridization) marked a crucial evolution in spatial transcriptomics. These methods utilize multiple fluorescent markers to enhance spatial resolution, enabling the detection of thousands of RNA species simultaneously within a single tissue Section. [75, 76]. MERFISH, in particular, has achieved subcellular resolution, allowing for precise localization of mRNA molecules within individual cells [77]. Recent innovations like Visium and Stereo-seq have further transformed spatial transcriptomics by integrating high-throughput sequencing with spatial barcoding techniques. Visium technology, developed by 10x Genomics, captures gene expression data from tissue sections while maintaining spatial context, achieving near single-cell resolution [78]. Conversely, Stereo-seq employs DNA nanoballs for RNA capture, demonstrating even higher resolution capabilities, with spot sizes as small as 220 nm [79]. These advancements enable comprehensive spatial transcriptomic maps that reveal intricate cellular interactions and tissue architecture.

A variety of computational methods, including graphical analysis and machine learning, have been developed to analyze ST data [78]. These methods aim to uncover complex biological relationships and regulatory mechanisms among cells. Researchers often employ various visualization tools post-analysis to illustrate the spatial distribution of gene expression, thereby adding to our understanding of gene localization within tissues and their association with specific cell types or pathological states. In 2016, Stahl et al. [80] reported an second-generation sequencing (NGS)-based ST approach that captures whole transcriptome information from spatially barcoded microarrays in tissue sections. This approach is capable of both preserving tissue spatial location information and obtaining transcriptional information from different tissue spatial loci. However, the method does not operate at the single-cell level. Recently, 10X Genomics released Visium spatial transcriptome sequencing, an improved version of this technology, which offers significantly enhanced resolution and sensitivity [81]. The integration of scRNA-seq data with ST data allows for the spatial localization of different single-cell subpopulations, annotation of spatial transcriptome regions by cell type, precise elucidation of cellular interactions, spatial mapping of specific cellular subpopulations in development and disease, and understanding the mechanisms by which these subpopulations synergistically form tissue phenotypes. The field of integrative computational methods has seen significant advancements due to progress in scRNA-seq and ST research (Table 1) [78, 82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106]. The kidney, one of the most diverse organs in mammals in terms of cell population and spatial structure, exhibits considerable variation in the spatial distribution of different cell types. Further spatial histological analyses are therefore necessary to determine the number and function of major cell subtypes and their characteristic proteins in greater depth. Such analyses will help reveal the spatial heterogeneity of this complex organ.

Table 1 Single-cell and Spatial integration strategies

ST in renal research

ST in the study of the Spatial heterogeneity of glomerular diseases

The kidney is recognized as one of the most diverse organs in mammals, exhibiting a complex cellular composition and spatial architecture. It comprises over 30 distinct functional cell types, organized into a highly compartmentalized structure that includes glomerular, tubular, and interstitial systems. This intricate spatial arrangement is crucial for sustaining normal renal function and homeostasis [107]. Conventional scRNA-seq has emerged as a powerful tool for exploring cellular heterogeneity within the kidney; however, it has notable limitations regarding the spatial organization and interactions among renal cells. Specifically, scRNA-seq does not provide precise information about the topological localization of specific cell subpopulations within the glomerulus, which is vital for understanding their roles in glomerular function and their responses to pathological conditions [108]. Furthermore, scRNA-seq fails to capture the spatial gradients of cell state transitions during injury and repair processes, thereby limiting insights into renal regeneration dynamics [109, 110]. This inability to elucidate the complex network of interactions between immune cells and renal parenchymal cells further complicates our understanding of renal homeostasis and injury responses [109]. The limitations of scRNA-seq highlight the urgent need for advanced spatially resolved transcriptomics approaches that integrate gene expression data with the spatial context of renal cellular architecture. Techniques such as Slide-seqV2 and other high-resolution spatial transcriptomics methods have demonstrated the potential to uncover disease-specific cell neighborhoods and pathways, thereby providing a more comprehensive understanding of kidney pathology [111, 112]. These methodologies allow for the mapping of cellular interactions and the identification of transcriptional signatures specific to various nephron segments, which is essential for elucidating the complex interplay of cells within the kidney [113]. In conclusion, ST technology offers a novel perspective for renal research by addressing the limitations of scRNA-seq, particularly in terms of spatial information. By resolving gene expression profiles while preserving information about cellular spatial locations, ST technology enhances our understanding of the cellular composition and functional states across different renal regions. This spatial resolution is crucial for elucidating changes during disease progression, thereby offering deeper insights into the complex mechanisms underlying renal diseases.

Spatial transcriptomics, as an advanced tool for elucidating the molecular mechanisms of complex diseases, has shown significant promise in studying glomerular diseases. By integrating gene silencing, pathway activation, and precise targeting, Space Deleted Polypeptide Technology (DSP technology) has successfully identified 3,205 glomerular-specific gene expression alterations in chronic glomerulopathies, such as systemic sclerosis and membranous glomerulopathies. Notably, this analysis revealed that certain up-regulated genes are involved in collagen metabolism and cell adhesion pathways, while down-regulated genes are associated with NOTCH signaling regulation and DNA methylation. These findings not only illuminate common molecular pathways in disease progression but also lay a theoretical foundation for clinical target identification [108]. In C3 glomerulopathy, related to the complement system, spatial transcriptomics localized the aberrant expression of complement regulators, such as CFH and CFI, clarifying the mechanisms behind local complement pathway dysregulation [113]. Additionally, studies on IgA nephropathy utilized spatial transcriptome analysis to reveal the molecular basis of inflammatory and oxidative stress pathways, correlated with the extent of thylakoid proliferation in pathologic staging [114]. Such insights not only uncover critical molecular features of glomerular diseases but also bolster theoretical support for precision diagnosis and targeted treatments.

From a regional and cell-cell interaction perspective, significant transcriptomic changes have been identified in diabetic kidney disease (DKD) studies, highlighting spatial differences between glomeruli and tubules that suggest key targets for filtration barrier damage. By integrating scRNA-seq with spatial localization analyses, researchers have elucidated ligand-receptor regulatory mechanisms among podocytes, endothelial cells, and various immune cells within the glomerular microenvironment, thereby providing insights into local inflammatory response regulation [109, 112]. In the case of lupus nephritis, spatial transcriptomics revealed that B cells form localized expansion centers and differentiate into plasma cells, suggesting new therapeutic approaches, such as anti-CD20 antibody therapy, and enhancing our understanding of the disease’s complexity [115].

Regarding the exploration of diagnostic and therapeutic targets, high-resolution localization technologies, including NanoString and Visium, offer innovative methods for identifying molecular markers relevant to disease staging by conducting single glomerular-scale genotypic analyses in conjunction with histopathological features, such as the degree of fibrosis [108]. Algorithms like Giotto and SPOTlight further enable the isolation and validation of the spatial localization of rare monocytes within the glomerulus, differentiating them from mixed signals [116]. In terms of therapeutic targets, spatial transcriptomics has revealed deficiencies in CD46 expression in glomerular epithelial cells related to complement regulator mutations in C3G. This finding underscores the potential for targeted therapy using localized complement inhibitors, such as anti-C5 antibodies [113]. Furthermore, in lupus nephritis models, the observation that APOE + monocytes mediate glomerular fibrosis via TGF-β signaling emphasizes the need for combined immunomodulatory and antifibrotic interventions [116].

In summary, spatial transcriptomics provides a powerful tool for uncovering disease-specific molecular networks, cellular interactions, and spatial heterogeneity by preserving in situ gene expression information within glomerular structures. This approach opens new avenues for precision diagnosis and targeted therapy. Future research should focus on integrating larger sample cohorts, cross-species validation, and dynamic time series analyses to further enhance its applications in addressing glomerular diseases.

ST in the study of the mechanisms of renal tubular injury

Spatial transcriptomics technology has demonstrated significant application value in the study of renal tubular injury, providing new perspectives to reveal the molecular mechanism of renal tubular injury by integrating high-resolution spatial location information with global gene expression profiles. For instance, studies utilizing the 10x Genomics ST platform on both murine and human renal cortices have uncovered significant variability in the intercellular interaction networks, especially between proximal tubules and peripheral vascular regions [111]. This variability provides critical insights into the functional compartmentalization of the renal tubules, which is vital for understanding kidney function.

In models of acute kidney injury (AKI), spatial transcriptomics has elucidated region-specific gene expression alterations in renal tubular epithelial cells. Notably, ischemia-reperfusion injury triggers an elevated expression of Havcr1 specifically in the outer medulla. Conversely, Lcn2 is upregulated in both the cortex and medulla, highlighting distinct injury response patterns among different segments of the renal tubule [109]. Additionally, studies on chronic kidney disease (CKD) have pinpointed that the regional deletion of TGF-β signaling in proximal tubules leads to heightened mitochondrial damage and inflammation, underscoring the pathway’s crucial role in maintaining renal tubular health [117].

In both AKI and CKD contexts, metabolic reprogramming and intercellular communication have emerged as pivotal processes in renal tubular repair and fibrosis. In cases of AKI, genes associated with fatty acid metabolism show increased expression in the renal cortex, potentially indicating a metabolic adaptation to hypoxic conditions [109]. Meanwhile, in CKD models, proximal tubular cells appear to communicate with distal tubular and endothelial cells via the NRG3-ERBB4 ligand-receptor axis, and this paracrine signaling intensifies in pathological states, presenting a new angle on tubular intercellular signaling dynamics [118].

The spatial heterogeneity of regulatory networks involved in fibrosis and inflammation has also garnered attention. Research highlights that deficiencies in TGF-β signaling lead to the upregulation of extracellular matrix deposition genes, such as collagen synthase, indicating a spatially localized function for this pathway in tubulointerstitial fibrosis [108]. Furthermore, activation of the WNT/β-catenin pathway was observed to co-localize with pro-inflammatory markers in diabetic kidney disease (DKD), suggesting a synergistic relationship between inflammation and fibrosis that varies spatially within the kidney [119].

The advent of high-resolution spatial transcriptomic technologies, such as Slide-seqV2 and MERFISH, has propelled research to the single-cell level. These innovations not only clarify the roles of rare pro-fibrotic progenitor cells but also facilitate the dynamic tracking of cellular interaction networks during renal injury and repair [112]. When combined with in vivo imaging and spatial histology techniques, future studies are poised to uncover the spatiotemporal dynamics of the renal microenvironment and its regulatory mechanisms involved in disease processes. Overall, the advancements in spatial transcriptomics underscore its pivotal role in examining renal tubular injury and repair’s complex microenvironment. This technology lays a theoretical groundwork for developing targeted therapeutic strategies aimed at the cellular communication networks within the kidney, ultimately advancing the diagnosis and treatment of renal diseases to new heights.

ST in the study of renal tumors

The application of spatial transcriptomics in renal tumor research has made remarkable strides in recent years. By preserving the spatial information of gene expression, this innovative technology offers fresh insights into understanding the tumor microenvironment, cellular heterogeneity, and metabolic characteristics. Its significance is particularly evident in advancing the study of renal clear cell carcinoma (ccRCC). Research has shown that ccRCC tumors exhibit distinct functional and metabolic differences between their central and peripheral regions. The central region is marked by heightened glycolysis and oxidative phosphorylation, while the marginal or peripheral region demonstrates enhanced angiogenesis and epithelial-mesenchymal transition (EMT)-related pathways, mechanisms that may drive tumor invasion [120]. Notably, fatty acid metabolism appears to be more pronounced in areas closer to the tumor center, accompanied by elevated succinate accumulation. Conversely, purine metabolic activity is amplified in borderline regions, suggesting a potential metabolic gradient that could influence therapeutic resistance [121]. Furthermore, spatial transcriptomics has shed light on cellular interactions and heterogeneity within tumors. It has successfully mapped the co-localization patterns of cancer-associated fibroblasts (CAFs), immune cells such as T cells, and tumor cells while also capturing their functional states [120]. These findings have unveiled potential interaction mechanisms and provided a theoretical foundation for developing targeted therapeutic strategies. These discoveries underscore the transformative potential of precision medicine informed by spatial transcriptomics data. By selectively targeting key metabolic nodes, such as fatty acid synthase, this approach may pave the way for optimized therapies that specifically address the metabolic heterogeneity of tumors [122]. In summary, spatial transcriptomics represents a multidimensional research framework for exploring the complex ecosystem of renal tumors, offering broad prospects for uncovering molecular mechanisms and refining diagnostic and therapeutic protocols.

Current major challenges for spatial transcriptomics include improving resolution, reducing costs, streamlining the data analysis process, and extending applicability to more tissue types. Future directions may include the development of higher-throughput ST technologies to enable larger-scale tissue analysis; and the integration of multi-omics data, such as proteomics and metabolomics, to provide a more comprehensive view of tissue function. In addition, as technology advances, ST technologies are expected to play a greater role in disease diagnosis, treatment response monitoring, and personalized medicine.

Single-cell atlas of the human kidney using single-cell and ST techniques

Fig. 4

Single-cell analysis strategies for pathological kidney samples. A visual summary of the methodologies employed for single-cell analysis in pathological kidney samples: scRNA-seq, snRNA-seq and spatial transcriptomics

In recent years, significant progress has been made in combining single-cell sequencing and ST data to explore the overall molecular map of kidney development and disease progression (Figs. 3 and 4). The nephron is the basic functional unit embedded in the mesenchyme and maintains kidney homeostasis and participates in disease development. It consists of endothelial cells, epithelial cells, mesangial cells, fibroblasts, myofibroblasts, as well as immune cells [123]. The most diversified types and spatial distribution of cells in the kidney pose a major challenge when studying renal physiology using scRNA-seq. Specific sequencing strategies and bioinformatics analysis have therefore been developed to integrate individual cell RNA data with spatial information (Table 1), providing new insights into the kidney. Stewart et al. [124] combined scRNA-seq with mass spectrometric analysis to analyze the anatomical location of immune cells in the human kidney and described changes in immune cell subsets in different anatomical regions over time under external stimuli. The study showed that multiple immune cell subsets are formed during the first three months of human fetal kidney development. However, the transcription profiling of these cells was different from that of immune cells in the mature kidney. Further evidence from the Bulk transcriptome and a mouse model of pyelonephritis confirmed that epithelial cells may be involved directly in the defense against kidney infection after birth by secreting antimicrobial peptides. Trajectory analysis results demonstrated a temporal progression from the fetal M1-macrophages transcriptome to mononuclear cells in the fetal kidney, with different expression of the pro-inflammatory M1 gene and lower enrichment of genes for antigen processing and presentation in dendritic cells (DC) cells. A marked asymmetrical distribution of immune cells in the kidney was also observed, with B cells located mainly in the cortex and mononuclear phagocytes enriched in the deeper regions. Giotto SPOTlight, an open-source computing tool, was used by Dixon et al. [81] to optimize and validate the spatial map of gene expression and the cell-cell interaction network in bilateral ischemia-reperfusion injury and repair in mice kidney. Ferreira et al. [109] optimized ST and mononuclear sequencing datasets to predict cell types in human renal tissue specimens after nephrectomy, and showed consistency with underlying histopathological changes in the spatial transcriptome characteristics of models of AKI. Similarly, Janosevic et al. [125] used single-cell sequencing in conjunction with spatial transcriptome sequencing to analyze the molecular characteristics of different segments of nephrons during progression of sepsis disease. Li et al. [126] also mapped the single-cell transcriptome of intratumoral and associated regions of renal cancer using single-cell plus spatial transcriptome sequencing. This approach has the potential to providing important information and a theoretical basis for the diagnosis and treatment of kidney diseases, thereby contributing to a deeper understanding of renal pathophysiological processes and the development of individualized therapeutic strategies. In conclusion, single-cell sequencing can aid in the study of renal spatial information and facilitate our comprehension of the distribution of various cell types within renal tissues and their interactions, with this information being crucial for furthering our understanding of the kidney’s structure and function, and the mechanisms underlying related diseases.

Future perspectives

Technological advancements

Single-cell sequencing and spatial transcriptomics are poised to revolutionize renal research by entering an “era of multidimensional dissection.” Enhanced sensitivity of snRNA-seq and subcellular spatial resolution will become standard tools in renal studies [127]. Multi-omics integration will enable comprehensive profiling from gene expression to metabolic states, while automated microfluidic platforms with AI-driven design will improve throughput and reproducibility [128]. Next-generation spatial transcriptomics platforms may achieve single-cell resolution and incorporate epigenetic marker detection. Quantum biotechnology and nanoscale sensors could enable real-time, non-destructive multimodal data acquisition, including mechanosignals. AI-powered virtual cell modeling will integrate single-cell data with imaging to construct dynamic renal ecosystem maps [129].

Scientific breakthroughs

Single-cell technologies will elucidate the role of cellular continuum states in renal fibrosis. Integrating scRNA-seq with spatial metabolomics will precisely delineate metabolic reprogramming pathways during EMT. Spatiotemporal atlases will elucidate regulatory networks of cell fate determination during kidney development [130]. Holistic renal deconvolution will integrate single-cell multi-omics data with biophysical models to simulate renal functions from molecular to organ levels.

Clinical translation

Single-cell-guided precision medicine will transform renal clinical practice. Spatial transcriptomics-based molecular subtyping could predict responses to therapies in chronic kidney disease [108]. Single-cell drug resistance signatures will optimize treatment regimens [131]. Dynamic monitoring of renal cells through single-cell profiling may enable real-time risk assessment of renal disease progression. Regenerative medicine may see breakthroughs in renal tissue engineering through single-cell lineage tracing and spatial microenvironment reconstruction [132]. AI-assisted diagnostic systems integrating single-cell data with electronic health records will establish predictive models for renal disease progression.

Challenges and ethical considerations

Technological advancements must address data sovereignty and biosecurity challenges. Risks of privacy breaches through single-cell-level identification require robust mitigation strategies. The convergence of synthetic biology and single-cell editing technologies demands ethical governance frameworks to ensure responsible use in renal research and clinical applications [133, 134].